Multiple paternity in two populations of finetooth sharks (Carcharhinus isodon) with varying reproductive periodicity

Abstract The mechanisms underlying polyandry and female mate choice in certain taxonomic groups remain widely debated. In elasmobranchs, several species have shown varying rates of polyandry based on genetic studies of multiple paternity (MP). We investigated MP in the finetooth shark, Carcharhinus isodon, in order to directly test the encounter rate hypothesis (ERH), which predicts that MP is a result of the frequency of encounters between mature conspecifics during the breeding season, and should therefore increase when more time is available for copulation and sperm storage. Female finetooth sharks in the northern Gulf of Mexico (GoM) have been found to reproduce with both annual periodicity and biennial periodicity, while finetooth sharks from the northwestern Atlantic Ocean have only been found to reproduce biennially, allowing us to compare mating opportunity to frequency of MP. Our results show high rates of MP with no significant difference in frequency between females in the GoM (83.0%) and Atlantic (88.2%, p = .8718) and varying but nonsignificant rates of MP between females in the GoM reproducing annually (93.0%) and biennially (76.6%, p = .2760). While the ERH is not supported by this study, it remains possible that reproductive periodicity and other physiological factors play a role in determining rates of MP in elasmobranchs, with potential benefits to individuals and populations.

have a better chance of recovering from population decline, though it has also been suggested that polyandry may actually lower genetic diversity and N e by increasing the variance in male reproductive success (Karl, 2008;Lotterhos, 2011).
The advancement of genetic techniques has made studying female mating strategies in nature more feasible because of the ability to detect polyandry via multiple paternity (when a single brood is sired by more than one male) using DNA analysis of the female and offspring . Molecular studies in the past 20 years have revealed a surprising amount of polyandry in systems that were expected to be genetically monogamous, especially where social monogamy was observed, such as in passerine birds (Petrie & Kempenaers, 1998) and mammals (Thonhauser et al., 2013), where females invest heavily in reproduction and are therefore expected to derive little benefit from remating (Zeh & Zeh, 2001). Recently, the genetic mating systems of elasmobranchs (sharks, skates, and rays) have garnered interest because of growing concern for shark population persistence and conservation (Dulvy et al., 2014).
Elasmobranchs have a relatively wide variety of reproductive strategies for a large vertebrate, ranging from oviparity (egg-laying) to several forms of viviparity (live birth), including placental viviparity (Parsons et al., 2007;Pratt & Carrier, 2001). All elasmobranchs use internal fertilization, and females invest heavily in reproduction compared with males through long gestation periods (4.5-36 months) and large, energetically expensive young (Conrath & Musick, 2012;Tanaka et al., 1990). Yet, polyandry leading to multiple paternity has been shown to be highly common in elasmobranchs and thought to be facilitated by female sperm storage (Fitzpatrick et al., 2012), which is common and can persist in some species for years (Conrath & Musick, 2012;Pratt & Carrier, 2001). Although frequency of multiple paternity has been shown to vary widely both within and between species in a way that implies local adaptation (Chabot & Haggin, 2014;Daly-Engel et al., 2007Fitzpatrick et al., 2012), the direct behavioral causes and potential ultimate evolutionary benefits of this behavior in sharks remain unclear.
Several hypotheses attempt to explain the adaptive significance of multiple paternity in sharks (reviewed in Fitzpatrick et al., 2012).
Multiple mating has obvious benefits to male fitness since males will likely sire more offspring with each additional mate. Conversely, polyandrous mating behavior by female sharks may actually decrease their fitness as a result of wounds inflicted during copulation, when males are known to grasp the flanks and pectoral fins of the females with their teeth (Pratt & Carrier, 2001). As a result, the study of multiple mating in sharks has long focused on the role of females and female choice (Daly-Engel et al., 2010;Fitzpatrick et al., 2012). Multiple paternity may be favored to evolve in species in which the female has a lower risk of injury if she submits to copulation, a hypothesis known as convenience polyandry (DiBattista et al., 2008). Alternatively, polyandry may be a function of female mate choice, either pre-or postcopulatory, and ultimately increase survival of young by increasing the chance of fertilization by a high-quality or genetically compatible male (Watson, 1991;Yasui & Garcia-Gonzalez, 2016;Zeh & Zeh, 2001). The simplest hypothesis proposed, and the one we test in the current study, is the encounter rate hypothesis (ERH), which states that the rate of multiple mating increases with an increase in frequency of encounters between mature males and receptive females during the mating season (Boomer et al., 2013;Daly-Engel et al., 2006Nosal et al., 2013).
According to the ERH, the more time available for mating between broods, the higher the predicted rate of multiple paternity.
We tested the ERH by estimating multiple paternity rates in two populations of the finetooth shark (Carcharhinus isodon; Figure 1), a small coastal requiem shark that uses placental viviparity to reproduce (Castro, 1993;Compagno et al., 2005). The most recent U.S. stock assessment, conducted in 2007, characterizes the finetooth shark in the northwestern Atlantic Ocean and Gulf of Mexico (GoM) as a single stock (NOAA, 2007), though genetic analysis has shown significant population structure between Atlantic and GoM populations, indicating little historical migration between ocean basins (Portnoy et al., 2016). Population-level differences abound between these regions; as a result of warmer waters in the GoM, finetooth sharks in the GoM do not migrate as far as conspecifics on the Atlantic coast (Castro, 1993;Driggers & Hoffmayer, 2009;Drymon et al., 2010), while finetooth sharks mature more slowly in the northwestern Atlantic than they do in the GoM (Higgs et al., 2020;Vinyard et al., 2019), potentially facilitating differences in fecundity, growth rate, and mating strategy (Carlson et al., 2003).
As with many members of this speciose genus, parturition in the finetooth shark occurs in early spring after a gestation period of approximately one year and is thought to be followed by a "resting year" in which the animal does not reproduce (Brown et al., 2020;Castro, 1993;Higgs et al., 2020). In a study by Castro The authors postulated that some sharks may switch from biennial to annual reproduction as a result of energy allocation; full-grown individuals with adequate food supply that are not required to migrate long distances could allocate that energy to increasing their frequency of reproduction. Since then, a study with a larger sample of females (n = 50) captured during the peak ovulation/parturition period of (April-June) confirmed the presence of both reproductive periodicities in the northern GoM, showing approximately a 65:35 ratio (32 annual, 18 biennial) among females (Higgs et al., 2020).
Because such divergence in reproductive periodicities is unusual among sharks (Driggers & Hoffmayer, 2009), this species represents a rare opportunity to examine the ecological dependence and local adaptive value of multiple paternity, including potential effects on population genetic diversity.
In this study, we compared rates of multiple paternity and standing gene diversity (including allelic richness) between two populations of sharks with varying reproductive periodicities using 12 highly polymorphic microsatellite DNA loci. Little is directly known about sperm storage and competition in most sharks, including the finetooth shark, but females from many shark species are capable of storing sperm in the oviducal gland for months to years before eggs are fertilized (Pratt, 1993), and sperm competition in response to polyandry has been shown to drive selection on reproductive traits in sharks and bony fishes Rowley, Locatello, et al., 2019).
Given the ubiquity of multiple paternity in elasmobranchs in general and the genus Carcharhinus in particular (Byrne & Avise, 2012;Fitzpatrick et al., 2012), plus the fact that female finetooth sharks can likely store sperm from multiple conspecifics and in the absence of data on sperm competition, we assume here that any male they mate with could genetically contribute to the next litter. Finetooth sharks aggregate and mate in May-June and have a gestation period of 11-12 months (Castro, 1993;Higgs et al., 2020), so we assume that under the ERH, a female shark with biennial reproduction could participate in twice as many mating periods between litters compared with an annual reproducer, resulting in a correspondingly higher rate of multiple paternity. We therefore hypothesize that the overall frequency of MP would be lower for finetooth shark females in the GoM, where some individuals are annual reproducers, and higher in the Atlantic, where all individuals are biennial reproducers. We further predict that MP will be lower among annual reproducers in the GoM compared with biennial conspecifics in the same population, which could result in a loss of diversity in the region with lower MP. following the methods described in Higgs et al. (2020) in females caught during the peak mating/parturition period, with females exhibiting simultaneous vitellogenesis and gestation being classified as reproducing annually. Small (~1 cm 3 ) fin or muscle samples were obtained from the gravid females and each in utero pup and were stored in 1.5 ml DMSO buffer or >75% ethanol. DNA extraction was done via a salting-out procedure adapted from Sunnucks and Hales (1996). The genetic mating system of the finetooth shark was assessed using adult female DNA from 98 GoM and 30

| MATERIAL S AND ME THODS
Atlantic specimens. A suite of microsatellite markers specific to finetooth sharks (Giresi et al., 2012a)  Alleles in mothers and offspring were scored visually using GENEIOUS, and any litters with more than two nonmaternal alleles at two or more loci indicated the presence of multiple sires. Litters showing multiple paternity at only one locus were not included. This method of scoring serves as a conservative baseline because it assumes that every male in the population is a heterozygote at every locus, which is unlikely in elasmobranchs, which have naturally low rates of molecular evolution compared with other taxa (Martin et al., 1992). Additional methods for paternity analysis included the programs FMM (Frequency of Multiple Mating; , which uses Bayesian priors that account for population allele frequencies to generate a 95% confidence interval for frequency of MP, and PrDM (Probability of Detecting Multiple mating; , which calculated the power of our locus set to detect MP in litters of varying sizes and in the presence of paternal skew. We tested three paternal contribution scenarios: (a) two sires with even skew (0.5:0.5), (b) two sires with moderate skew (0.33:0.67), and (c) two sires with high skew (0.1:0.9). This is particularly important in this study because of the small mean litter size; the probability of detecting multiple mating is greatly increased when litter size increases, and it is likely that sperm competition plays a role in the mating strategy of the finetooth shark, similar to other sharks (Portnoy et al., 2007;Rowley, Locatello, et al., 2019). To avoid potential type II error, a "scaled MP" value was calculated for each population grouping based on mean litter size and paternal skew, in order to estimate the rate of MP if our probability of detection was 100%. Specifically, the scaled value for each population was calculated as the MP rate according to FMM divided by the PrDM value for the average litter size for that population, to reflect how much more multiple mating might be detected if PrDM = 1.
The program GERUD v.2.0 (Jones, 2005) was used to estimate the number of sires in a brood across all loci simultaneously by reconstructing parental genotypes from the genotypes of the progeny.
The program COLONY v.2.0.6.4 (Jones & Wang, 2010) was used to extract parentage and sibship information from genotype data using a full likelihood method. As these programs reconstruct paternal genotypes from the offspring data and assign parentage, they can also show whether reproductive skew has occurred using genotype reconstruction or maximum likelihood, respectively. Molecular indices of diversity among unrelated individuals were generated in FSTAT v1.2 (Goudet, 1995). These included Nei's (1987) unbiased estimator of gene diversity (H) and allelic richness (A r ), which provides a measure of allelism that is corrected for and independent of sample size, allowing for accurate comparison among groupings with unequal sampling. Paired t tests were used to determine whether differences in diversity metrics across loci were significant between the Atlantic and GoM, and between annual and biennial reproducers.
Fisher's exact test was performed using SAS Software, version 9.4 (SAS Institute, Inc., Cary, NC) to detect significant variation between groups. Statistical significance was defined a priori as p < .05.

| RE SULTS
The mean litter size was 4.02 (SD = 1.04) pups in the GoM and 4.06 (SD = 0.64) in the Atlantic, and brood size ranged between 1 and 9 individuals. To avoid extrapolation and ensure our ability to detect multiple paternity, only gravid females with three or more pups were analyzed (Daly-Engel et al., 2007). This included 92 of the 98 adults and their pups from the northern GoM (N = 481 individuals) and 17 of the 30 adults and their pups from the northwest Atlantic (N = 86 individuals). Reproductive periodicity was known for 34 of the females from the GoM (biennial N = 10, annual N = 24); all females in the Atlantic were reproducing biennially (N = 30).
No significant deviation from Hardy-Weinberg equilibrium and no evidence of genotyping errors, including null alleles, large allele dropout, or pipetting error, were found at the 12 loci used in this study (Table 1). With a mean litter size of approximately four pups, the PrDM output for four offspring provides the nearest estimation of the actual detection probability for the population overall. The average probability of detecting multiple paternity was estimated to be between 33% (for high skew) and 85% (for even skew), depending on the ratio of genetic contribution. COLONY and GERUD results indicated no evidence of paternal skew in any litter (e.g., litters of four pups with two sires primarily had a ratio of 2:2, and the largest litter of nine pups had three sires with a ratio of 2:3:4 pups per sire), so even (0.5:0.5) male contribution was assumed, for which the PrDM program gave a probability of detecting multiple mating of 85%.

The most conservative estimate of MP based on visual scoring
showed 53 out of 92 litters from the GoM having three or more paternal alleles at two or more loci, giving an overall estimated minimum frequency of MP of 57.6%. An additional 11 litters had three or more paternal alleles at only one locus, which was not considered sufficient to demonstrate MP under this method. The remaining 28 litters had no evidence of MP. In the Atlantic population, 10 out of 17 litters had three or more paternal alleles at two or more loci, giving a minimum expected frequency of MP of 58.8%. A further five litters had three or more paternal alleles at only one locus, and two litters had no evidence of multiple paternity. Within the GoM, frequency of MP based on visual scoring of annual litters was estimated to be 67% (16 of 24 litters), while frequency of MP among biennial litters was 70% (7 of 10 litters).   Table 2). Despite high polyandry, we found no evidence for increased fitness as a result of multiple mating; also, as in previous studies on sharks (Boomer et al., 2013;Daly-Engel et al., 2010;Portnoy et al., 2007), litter size and rate of MP were not correlated.

| D ISCUSS I ON
Though a predominance of polyandry was detected among female finetooth sharks, the frequency of MP varied somewhat between the different methods used in this study. Visual scoring is the most conservative method because it assumes that males are heterozygous at all loci, which may undercount MP at loci with low polymorphism, and may have accounted for the difference observed in this study. However, FMM takes into account the number of microsatellite loci, their degree of polymorphism, and the number of pups in each litter to provide a less conservative estimate of multiple mating. While these numbers differ, it is apparent that a majority of females in this population are polyandrous, with an above-average frequency compared with other shark species studied to date (for a review, see Rossouw et al., 2016). Furthermore, given the low probability of detecting multiple mating in smaller litters, our unscaled results are more likely an underestimation than an overestimation.
There are no documented observations of mating in finetooth sharks, but studies have shown that this species aggregates in large numbers (Castro, 1993). Castro (1993)  shown the occurrence of seasonal concentrations of adult finetooth in coastal waters (Parsons et al., 2007;Bethea et al., 2014). The ERH (Daly-Engel et al., 2007)  can also increase MP (Daly-Engel et al., 2007). Mobbing or crowding, which occurs when multiple males simultaneously attempt to coerce mating with a single female, has been observed in nurse sharks (Pratt & Carrier, 2001) and whitetip reef sharks (Whitney et al., 2004 the western Indian Ocean (Pirog et al., 2020). Support for increased multiple mating among organisms that aggregate more densely can also be found among other taxa, such as sea turtles (for a review, see Lee et al., 2018  . The GoM population of finetooth sharks showed no evidence for reproductive skew, which could indicate a lack of long-term sperm storage and/or postcopulatory sexual selection. However, it is possible that females mating on an annual cycle have little to no paternal skew due to a lack of time for sperm competition to occur. Females on a biennial cycle that do not take a resting year, on the other hand, could mate immediately following parturition and store sperm throughout their postpartum year, allowing more time for postcopulatory sexual selection (Fitzpatrick et al., 2012;Rowley, Locatello, et al., 2019).

| CON CLUS ION
As marine apex and mesopredators, sharks are one of the most important taxonomic groups in marine ecosystems, helping to maintain the diversity of ocean habitats from shallow coastal waters to the deep sea. Because multiple mating appears to be common in elasmobranchs, there may be some ultimate genetic benefit that allows for increased adaptability in these long-lived vertebrates.
In general, understanding mating systems is crucial for estimating population viability in elasmobranchs with varying life-history characters, as many populations are faced with local decline. For finetooth sharks in particular, we recommend further investigation into the mechanisms that cause variation in reproductive periodicity, which may affect other aspects of shark mating systems. The current federal fishery regulations in the United States determine commercial and recreational take limits for all small coastal shark species, including finetooth sharks in both the GoM and coastal Atlantic (NOAA, 2007). However, marked differences in reproductive periodicity and genetic diversity, among other traits, indicate that each population may require separate management.

ACK N OWLED G M ENTS
The authors extend their thanks to the following: writing-review and editing (equal).

DATA AVA I L A B I L I T Y S TAT E M E N T
All primers and analyses used in this study have previously been published and are publicly available.